Install Cuda 4.0 on Ubuntu 11.04

NVIDIA Cuda 4.0 was released in this month. It includes Performance Primitives (NPP) library for image/video processing using multi-GPU programming. It is supported on many PCs and laptops with NVIDIA GPUs, including Thinkpad T410 with NVIDIA NVS 3100M.

Ubuntu 11.04 was released in the last month. Here is my instruction to install Cuda 4.0 on Ubuntu 11.04 on my Thinkpad T410 (thanks to instructions from http://forums.nvidia.com/index.php?showtopic=198030)

1. Install dependencies

sudo apt-get install build-essential gcc-4.4 g++-4.4 libxi-dev libxmu-dev freeglut3-dev

2. And download and install

• CUDA Toolkit for Ubuntu Linux 10.10 (sudo sh ...)
• CUDA Tools SDK (sudo sh ...)
• GPU Computing SDK code samples (no sudo)

from the following links:

LINUX	DOWNLOADS
Developer Drivers for Linux (270.41.19)	32-bit 64-bit
CUDA Toolkit C/C++ compiler CUDA-GDB debugger Visual Profiler GPU-accelerated BLAS library GPU-accelerated FFT library GPU-accelerated Sparse Matrix library GPU-accelerated RNG library Additional tools and documentation	documentation
CUDA Toolkit for Fedora 13	32-bit 64-bit
CUDA Toolkit for RedHat Enterprise Linux 6.0	64-bit
CUDA Toolkit for RedHat Enterprise Linux 5.5	32-bit 64-bit
CUDA Toolkit for RedHat Enterprise Linux 4.8	64-bit
CUDA Toolkit for Ubuntu Linux 10.10	32-bit 64-bit
CUDA Toolkit for OpenSUSE 11.2	32-bit 64-bit
CUDA Toolkit for SUSE Linux Enterprise Server 11 SP1	32-bit 64-bit
CUDA Tools SDK	32-bit 64-bit
GPU Computing SDK code samples	download

3. Download the Developer Drivers for Linux (270.41.19) from the above link. For driver installation, you need to

(1) remove the current nvidia packages you might have installed

sudo apt-get remove nvidia-common nvidia-current

(2) go to the text window via ctrl+alt+F2.

(3) stop X

sudo service gdm stop

(4) remove the nouveau module

sudo apt-get --purge remove xserver-xorg-video-nouveau

(5) stop gdm and restart X

sudo service gdm start

Restart the system

(6) stop X

sudo service gdm stop

(7) install driver

sudo sh devdriver_4.0_linux_32_270.41.19.run

(7) restart X

sudo service gdm start

4. Assuming you have accepted the default install location /usr/local/cuda, you should add the following in the .bashrc file of your home folder.

export CUDA_HOME="/usr/local/cuda"

export LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:${CUDA_HOME}/lib64"

export PATH=${CUDA_HOME}/bin:${PATH}

run ldconfig as root for it to take effect:

source ~/.bashrc

and

sudo ldconfig

5. make nvcc use gcc/g++ version 4.4, for example, in home directory

mkdir gcc44
cd gcc4
ln -s /usr/bin/cpp-4.4 cpp
ln -s /usr/bin/gcc-4.4 gcc
ln -s /usr/bin/g++-4.4 g++

edit /usr/local/cuda/bin/nvcc.profile to look in the gcc44 directory (give it the full path)

compiler-bindir =/home/xxx/gcc44

6. run make in NVIDIA_GPU_Computing_SDK/C

cd ~/NVIDIA_GPU_Computing_SDK/C
make

7. verify CUDA 4.0 installation by running the deviceQuery sample in the NVIDIA_GPU_Computing_SDK/C

bin/linux/release/deviceQuery

my Thinkpad T410 shows

[deviceQuery] starting...
linux/release/deviceQuery Starting...

CUDA Device Query (Runtime API) version (CUDART static linking)

Found 1 CUDA Capable device(s)

Device 0: "NVS 3100M"
CUDA Driver Version / Runtime Version 4.0 / 4.0
CUDA Capability Major/Minor version number: 1.2
Total amount of global memory: 256 MBytes (268107776 bytes)
( 2) Multiprocessors x ( 8) CUDA Cores/MP: 16 CUDA Cores
GPU Clock Speed: 1.47 GHz
Memory Clock rate: 810.00 Mhz
Memory Bus Width: 64-bit
Max Texture Dimension Size (x,y,z) 1D=(8192), 2D=(65536,32768), 3D=(2048,2048,2048)
Max Layered Texture Size (dim) x layers 1D=(8192) x 512, 2D=(8192,8192) x 512
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 16384 bytes
Total number of registers available per block: 16384
Warp size: 32
Maximum number of threads per block: 512
Maximum sizes of each dimension of a block: 512 x 512 x 64
Maximum sizes of each dimension of a grid: 65535 x 65535 x 1
Maximum memory pitch: 2147483647 bytes
Texture alignment: 256 bytes
Concurrent copy and execution: Yes with 1 copy engine(s)
Run time limit on kernels: Yes
Integrated GPU sharing Host Memory: No
Support host page-locked memory mapping: Yes
Concurrent kernel execution: No
Alignment requirement for Surfaces: Yes
Device has ECC support enabled: No
Device is using TCC driver mode: No
Device supports Unified Addressing (UVA): No
Device PCI Bus ID / PCI location ID: 1 / 0
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 4.0, CUDA Runtime Version = 4.0, NumDevs = 1, Device = NVS 3100M
[deviceQuery] test results...
PASSED

Press ENTER to exit...